fix: Launch now confirms when Unity is ready or restarted

[hatayama]

Summary

• uloop launch now waits until Unity can answer tool requests even when the Editor was already running.
• uloop launch -r now reports a structured restart result with previous and current process IDs.

User Impact

• Before this change, agents could only see a successful already-running launch as plain text, and restart success did not show whether a restart actually happened.
• After this change, agents can read JSON fields such as AlreadyRunning, Launched, Restarted, PreviousProcessId, and CurrentProcessId to confirm Unity is ready.

Changes

• Return structured launch, restart, and quit status payloads.
• Regenerate launch skill docs to document the machine-readable fields.
• Bump the CLI contract and Unity minimum CLI version for the new launch response contract.

Verification

• scripts/check-go-cli.sh
• cli/dist/darwin-arm64/uloop compile --project-path "$(git rev-parse --show-toplevel)"
• cli/dist/darwin-arm64/uloop launch --project-path "$(git rev-parse --show-toplevel)"
• cli/dist/darwin-arm64/uloop launch -r --project-path "$(git rev-parse --show-toplevel)"
• cli/dist/darwin-arm64/uloop execute-dynamic-code --project-path "$(git rev-parse --show-toplevel)" --code "return 12 + 12.22;"
• CLI_MINIMUM_VERSION_BASE_REF=origin/v3-beta CLI_MINIMUM_VERSION_HEAD_REF=HEAD scripts/check-cli-minimum-version-warning.sh
• git diff --check
• codex-review --parallel-tests "scripts/check-go-cli.sh" v3-beta

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 95f20519-d62f-44a2-968f-4105b72564da

📥 Commits

Reviewing files that changed from the base of the PR and between f145f48 and 2e97564.

📒 Files selected for processing (3)

• .agents/skills/uloop-launch/SKILL.md
• .claude/skills/uloop-launch/SKILL.md
• Packages/src/Editor/CliOnlyTools~/Launch/Skill/SKILL.md

✅ Files skipped from review due to trivial changes (1)

• .claude/skills/uloop-launch/SKILL.md

🚧 Files skipped from review as they are similar to previous changes (1)

• .agents/skills/uloop-launch/SKILL.md

---

📝 Walkthrough

Walkthrough

Bump CLI/tool version to 3.0.0-beta.28. Extend uloop launch to wait for tool readiness when reusing an existing Unity process, capture previous/current process IDs, and emit richer structured JSON responses for already-running, launched, restarted, and quit paths. Add tests and update SKILL docs.

Changes

Launch Command Refactoring

Layer / File(s)	Summary
Version contracts Packages/src/Editor/Domain/CliConstants.cs, cli/contract.json, cli/internal/tools/default-tools.json	CLI and tool metadata bumped from 3.0.0-beta.27 → 3.0.0-beta.28.
Launch response schema and serialization cli/internal/cli/launch_ready.go	launchReadyResponse extended with AlreadyRunning, Launched, Restarted, Quit, PreviousProcessId, CurrentProcessId, and ProjectRoot. Replaced single response writer with writeExistingLaunchReadyResponse, writeLaunchedReadyResponse, writeLaunchQuitResponse, all delegating to writeLaunchResponse for JSON serialization and error classification.
Launch command implementation and test coverage cli/internal/cli/launch.go, cli/internal/cli/launch_test.go	Launch flow now uses a spinner and waitForToolReadinessForLaunch when reusing an existing process, captures currentPid and optional previousPid, handles restart/quit via killUnityProcessForLaunch, and returns structured responses. Tests added to decode stdout JSON and assert flags and PIDs; existing focus tests updated to stub readiness probing.
User-facing skill documentation .agents/skills/uloop-launch/SKILL.md, .claude/skills/uloop-launch/SKILL.md, Packages/src/Editor/CliOnlyTools~/Launch/Skill/SKILL.md	Docs updated to explain focus vs launch/restart waiting behavior and the expanded JSON success schema including process-state flags and PID fields.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• hatayama/unity-cli-loop#1300: Updates the same cli/internal/cli/launch.go and cli/internal/cli/launch_ready.go structured response and readiness/wait flow.
• hatayama/unity-cli-loop#1027: Modifies uloop launch readiness/wait behavior and related readiness polling tests.
• hatayama/unity-cli-loop#1215: Changes the Unity-already-running focus workflow and tests touching cli/internal/cli/launch.go.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 14.29% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately reflects the main change: uloop launch now waits for Unity readiness and reports structured restart status with process IDs.
Description check	✅ Passed	The description is directly related to the changeset, clearly explaining the summary, user impact, changes made, and verification steps.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feature/launch-restart-status

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@Packages/src/Editor/CliOnlyTools`~/Launch/Skill/SKILL.md:
- Around line 55-66: The launch JSON schema in SKILL.md is missing the Quit
field emitted by writeLaunchQuitResponse; update the enumerated output fields to
include `Quit`: a boolean indicating whether the command quit the running Unity
process (true when a process was stopped via quit, false otherwise), and place
it alongside the other top-level flags (e.g., near `Launched`/`Restarted`) so
consumers can detect quit outcomes without parsing `Message`; ensure the field
name exactly matches `Quit` and the description matches the boolean semantics
used by writeLaunchQuitResponse.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: a03b8474-7340-4996-a99e-196ff3800079

📥 Commits

Reviewing files that changed from the base of the PR and between a36e661 and f145f48.

📒 Files selected for processing (9)

• .agents/skills/uloop-launch/SKILL.md
• .claude/skills/uloop-launch/SKILL.md
• Packages/src/Editor/CliOnlyTools~/Launch/Skill/SKILL.md
• Packages/src/Editor/Domain/CliConstants.cs
• cli/contract.json
• cli/internal/cli/launch.go
• cli/internal/cli/launch_ready.go
• cli/internal/cli/launch_test.go
• cli/internal/tools/default-tools.json

[cubic-dev-ai]

3 issues found across 9 files

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>